P R IMA R Y R E S E A R CH A R T I C L E
Declining population trends of European mountain birds
Aleksi Lehikoinen1 | Lluís Brotons2,3,4 | John Calladine5 | Tommaso Campedelli6 |
Virginia Escandell7 | Jiri Flousek8 | Christoph Grueneberg9 | Fredrik Haas10 |
Sarah Harris11 | Sergi Herrando12 | Magne Husby13 | Frederic Jiguet14 |
John Atle Kålås15 | Åke Lindström10 | Romain Lorrillière14,16 | Blas Molina7 |
Clara Pladevall17 | Gianpiero Calvi6 | Thomas Sattler18 | Hans Schmid18 |
Päivi M. Sirkiä1 | Norbert Teufelbauer19 | Sven Trautmann9
1Finnish Museum of Natural History, University of Helsinki, Helsinki, Finland
2InForest Jru (CTFC‐CREAF), Solsona, Spain3CREAF, Bellaterra, Spain
4CSIC, Bellaterra, Spain
5British Trust for Ornithology (Scotland), University of Stirling, Stirling, Scotland
6Italian Common Breeding Bird monitoring programme, Parma, Italy
7Estudio y seguimiento de aves/SEO/BirdLife, C/Melquíades Biencinto, Madrid, Spain
8Krkonose National Park, Vrchlabi, Czech Republic
9Dachverband Deutscher Avifaunisten (DDA) e.V., Geschäftsstelle, Münster, Germany
10Department of Biology, Lund University, Lund, Sweden
11The British Trust for Ornithology, The Nunnery, Norfolk, UK
12Catalan Ornithological Institute. Natural History Museum of Barcelona, Barcelona, Spain
13Section of Science, Nord University, Levanger, Norway
14Centre d’Ecologie et des Sciences de la COnservation (CESCO UMR 7204, Museum National d’Histoire Naturelle, Paris, France
15Norwegian Institute for Nature Research, Trondheim, Norway
16Université Paris‐Sud, UMR 8079 Ecologie Systématique Evolution, Orsay, France
17Snow and Mountain Research Center of Andorra (CENMA) ‐ Andorran Research Institute (IEA), Sant Julià de Lòria, Principality of Andorra
18Swiss Ornithological Institute, Sempach, Switzerland
19BirdLife Österreich, Wien, Austria
Correspondence
Aleksi Lehikoinen, Finnish Museum of
Natural History, University of Helsinki,
Helsinki, Finland.
Email: [email protected]
Funding information
Joint Nature Conservation Committee;
Finnish Ministry of Environment; Italian
Ministry of Agricultural, Food and Forestry
Policies; The Norwegian Environment
Agency; Royal Society for the Protection of
Birds; Catalan government; Italian Ministry
of Agricultural and Forestry Policies;
Swedish Environmental Protection Agency;
Climate and Environment Ministry; Rural
Network; Academy of Finland, Grant/Award
Number: 275606; BTO; Scottish
Government; Scottish Natural Heritage;
Abstract
Mountain areas often hold special species communities, and they are high on the list
of conservation concern. Global warming and changes in human land use, such as
grazing pressure and afforestation, have been suggested to be major threats for bio-
diversity in the mountain areas, affecting species abundance and causing distribution
shifts towards mountaintops. Population shifts towards poles and mountaintops
have been documented in several areas, indicating that climate change is one of the
key drivers of species’ distribution changes. Despite the high conservation concern,
relatively little is known about the population trends of species in mountain areas
due to low accessibility and difficult working conditions. Thanks to the recent
improvement of bird monitoring schemes around Europe, we can here report a first
account of population trends of 44 bird species from four major European mountain
Received: 9 February 2018 | Revised: 16 October 2018 | Accepted: 5 November 2018
DOI: 10.1111/gcb.14522
Glob Change Biol. 2018;1–12. wileyonlinelibrary.com/journal/gcb © 2018 John Wiley & Sons Ltd | 1
National Trust for Scotland; Scottish
Ornithologists' Club; Scottish
Mountaineering Trust; AEB Trust
regions: Fennoscandia, UK upland, south‐western (Iberia) and south‐central moun-
tains (Alps), covering 12 countries. Overall, the mountain bird species declined signif-
icantly (−7%) during 2002–2014, which is similar to the declining rate in common
birds in Europe during the same period. Mountain specialists showed a significant
−10% decline in population numbers. The slope for mountain generalists was also
negative, but not significantly so. The slopes of specialists and generalists did not
differ from each other. Fennoscandian and Iberian populations were on average
declining, while in United Kingdom and Alps, trends were nonsignificant. Tempera-
ture change or migratory behaviour was not significantly associated with regional
population trends of species. Alpine habitats are highly vulnerable to climate change,
and this is certainly one of the main drivers of mountain bird population trends.
However, observed declines can also be partly linked with local land use practices.
More efforts should be undertaken to identify the causes of decline and to increase
conservation efforts for these populations.
K E YWORD S
afforestation, agriculture, alpine habitat, common bird monitoring, global warming, land use
changes, loss of biodiversity, mountains, population trend, upland
1 | INTRODUCTION
Human land use changes and a changing climate are the major
threats to biodiversity around the world (Root et al., 2003; Stephens
et al., 2016; Travis, 2003). Habitat loss, fragmentation and degrada-
tion have affected species distribution ranges and abundances (Bail-
lie, Hilton‐Taylor, & Stuart, 2004; Fahrig, 2003). Global warming has
shifted species distribution areas towards the poles and mountain-
tops (Chen, Hill, Ohlemüller, Roy, & Thomas, 2011; Maggini et al.,
2011). From a conservation point of view, it is, however, equally
important to understand the effects of climate change on population
densities that do not necessarily coincide with distributional changes
(Chamberlain & Fuller, 2001). In general, while populations of low-
land bird and butterfly species have been shown to change accord-
ing to climate change scenarios in Europe and North America (Breed,
Stichter, & Crone, 2013; Devictor et al., 2012; Lindström, Green,
Paulson, Smith, & Devictor, 2013; Stephens et al., 2016), the popula-
tion status of species in the mountain areas is generally poorly
known (Chamberlain et al., 2012; Scridel et al., 2018; but see Flou-
sek, Telenský, Hanzelka, & Reif, 2015; Lehikoinen, Green, Husby,
Kålås, & Lindström, 2014).
Mountain areas often hold special species communities and are
thus in the high‐priority list of conservation (Rodriguez‐Rodriguez,Bomhard, Butchart, & Forster, 2011). Furthermore, mountain species
have been suggested to be particularly vulnerable to climate change,
since it is generally more difficult for them to find new suitable habi-
tats towards the mountaintops (low habitat availability simply
because of orography, Gonzalez, Neilson, Lenihan, & Drapek, 2010;
Huntley, Green, Collingham, & Willis, 2007; Sekercioglu, Schneider,
Fay, & Loarie, 2008) or in other mountain ranges (low connectivity
between them, Sirami et al., 2016). The rise in temperature associ-
ated with global warming has been predicted to be two to three
times higher in the 21st century than recorded during the 20th cen-
tury (Nogués‐Bravo, Araújo, Errea, & Martinez‐Rica, 2007). In addi-
tion to climate change, mountain species, especially species breeding
in uppermost open alpine areas, are also threatened by human land
use changes such as altered grazing pressure, afforestation, increased
disturbance of recreational activities, pollution (nitrogen and acid
deposition) and their interactions (Arlettaz et al., 2007; Brambilla
et al., 2010; Britton & Fisher, 2007; Herrando et al., 2016; Ims &
Henden, 2012; van der Wal et al., 2003).
The use of biodiversity indicators has become an increasingly
common way to monitor changes in the environment (Butchart
et al., 2010; Gregory et al., 2005, 2008). Indicators, such as Biodiver-
sity Change Index (Normander et al., 2012), Living Planet Index (Col-
len et al., 2009), and Red List Index (Butchart et al., 2005), gather
large number of information into a single index value, which are easy
to understand not only by scientists, but also policymakers and the
public (Gregory et al., 2005). Recent advances in this research field
have produced, for example, continental indicators of farmland birds
and climate change (Gregory et al., 2005; Stephens et al., 2016), but
a continental indicator for mountain areas has been lacking. To pro-
duce such indicators, large and long‐term datasets are required.
From the practical side, monitoring the fate of mountain species
may be particularly demanding as mountain areas are often difficult
to access, the number of species sharply decrease with altitude
(Zbinden et al., 2010) and population densities of species are low
(Lehikoinen et al., 2014). Thanks to the recent improvements of the
national bird monitoring in Fennoscandia (Norway, Sweden and
2 | LEHIKOINEN ET AL.
Finland), with new schemes covering also the most remote mountain
areas, a first‐ever regional bird indicator for the Fennoscandian
mountain range was created by Lehikoinen et al., (2014). In this
study, we have analysed mountain bird trends at the continental
scale, with data from 11 different mountain ranges in Europe.
The aim of this work was (a) to investigate population trends of
the common bird species in Europe breeding on high‐altitude moun-
tain habitats, (b) to evaluate whether population trends differ
between species with different ecological characteristics, which may
add information on underlying causes of population changes, (c) to
produce the first continental‐scale biodiversity indicator for moun-
tain bird communities and (d) to establish four regional mountain
bird indicators. The continental indicator will show the overall situa-
tion, whereas the regional indicators will tell more about the local
conditions (Gregory et al., 2005).
Based on the assumption that climate and land use conditions
have negatively affected species inhabiting mountain habitats
(Arlettaz et al., 2007; Brambilla et al., 2010; Herrando et al., 2016;
Ims & Hender, 2012; Lehikoinen et al., 2014), we hypothesize that
mountain bird species, in general, are declining in numbers. Second,
we hypothesize that this decline would be stronger in mountain
specialists that only occur in mountain areas in our study sites,
whereas mountain generalists, which also can be found at lower
elevations, are doing better because of generally higher ecological
flexibility (Davey, Chamberlein, Newson, Noble, & Johnston, 2012;
Davey, Devictor, Jonzén, Lindström, & Smith, 2013; Gough et al.,
2015). Third, we predict that population trends of mountain species
can be influenced by the migration status of species. We hypothe-
size that long‐distance migrants will have fared relatively poorly, as
they displayed on average more negative population trends in
recent years across Europe—whatever the elevation—than resi-
dents and short‐distance migrants (Laaksonen & Lehikoinen, 2013;
Sanderson, Donald, Pain, Burfield, & van Bommel, 2006; Vickery
et al., 2014). An alternative hypothesis is that if a change in habitat
quality in the mountain areas has a negative impact on species
which are spending the longest time in the mountain areas, short‐distance migrants and resident species should have faced stronger
declines than long‐distance migratory species (Lehikoinen et al.,
2014). Last, we hypothesize that the decline in mountain birds is
stronger at northern latitudes than at southern latitudes because
temperature is expected to increase more in the north (Jacob
et al., 2014).
2 | MATERIALS AND METHODS
2.1 | Data collection
Mountain bird populations have been monitored in 11 different
mountain areas distributed in 12 countries, mainly within national
monitoring schemes on common breeding birds using mainly system-
atic sampling (Supporting Information Table S1). In the present study,
we analysed data from 2002 to 2014. The data collection covered
this period unless stated otherwise: Fennoscandia (Finland, Norway
and Sweden), UK upland (Britain and Northern Ireland), the Giant
Mountains (Czech Republic, 2002–2011), the Alps (Austria 2008–2012, France, Germany 2005–2012, Italy, Switzerland), Massif Cen-
tral (France), the Pyrenees (Andorra 2011–2012, France, Spain), theApennines (Italy), Spanish central mountains (Spain), Spanish Iberian
mountain system (Spain), Baetica mountain range (Spain 2003–2012), and Cantabria mountain range (Spain; Table 1). The local cen-
sus methods are explained in Supporting Information Table S1. Cen-
sus methodology differed between countries, but this will unlikely
introduce systematic bias into the derived trends (see, e.g., Gregory
et al., 2005; Lehikoinen et al., 2014; Stephens et al., 2016).
2.2 | Site and species selection
To get enough data to calculate trends for a larger set of species,
we lumped the 11 areas into four larger mountain regions:
Fennoscandia, UK upland, south‐western mountains (including Pyre-
nees and four Spanish mountain areas, hereafter called as “Iberia”)
and the south‐central mountains (including the Alps and the sur-
rounding smaller mountains: Giant Mountains, Massif Central and
the Apennines, hereafter called as “Alps,” Figure 1).
Before we could define which species to use in the study, we
needed to define “mountain” monitoring sites and species in each
region. Our aim was to target species that prefer open or semi‐openmountain habitats. These are mainly situated on the highest altitudes
of the mountains and are thus in the highest risk in terms of climate
change (Gonzalez et al., 2010). Since mountaintop populations have
limited places to move upwards, the expected population declines
should be strongest in high‐altitude habitats. Thus, we selected
mountain tundra, meadows, grasslands, bare rock, sparsely vegetated
areas, peat bogs and scrubland above certain altitude. We also
included the, often spatially adjacent, zones of mountain birch forest
TABLE 1 The number of study sites (mean, min and max during2002–2014) in 11 mountain areas distributed over four majormountain regions. In the Giant Mountains and the Apennines, thenumber of point count locations were transformed into sites dividingnumber of point stations by 15 (a typical number in point countroutes in Italia and the Czech Republic, Giant Mountains)
Mountain area Region Mean sites
Fennoscandia Fennoscandia 160 (60–256)
UK upland UK upland 99 (72–140)
Alps “Alps” 122 (88–155)
The Giant Mountains “Alps” 1 (0–2)
Massif Central “Alps” 1 (0–2)
Apennines “Alps” 20 (9–37)
Baetica mountain range “Iberia” 6 (0–10)
Cantabria mountain range “Iberia” 12 (4–17)
Central mountain system “Iberia” 24 (16–29)
Iberian mountain system “Iberia” 6 (5–7)
Pyrenees “Iberia” 23 (11–39)
LEHIKOINEN ET AL. | 3
and dwarf mountain pine (for simplicity, all the mentioned habitats
are generally referred to as “mountain habitat”). For latitudinal rea-
sons (and also exposure on the western seaboard) also the altitudes
where open mountain habitat occur varies and this needs to be
defined separately for each mountain range. Since some of the spe-
cies occur also outside the mountains—though we were only inter-
ested in the populations living in the mountain areas—we needed to
use habitat information to define mountain sites from each area. For
instance, due to the long north‐east–south‐west gradient (1,600 km)
of the Fennoscandian mountain area, mountain habitats vary in alti-
tude. For example, tundra is first found above 1,300 m altitude in
the south, but at sea level in the very north (Lehikoinen et al., 2014).
It should be noted though, that only four out of 289 Fennoscandian
sites were situated below 100 m of altitude. In the rest of the moun-
tain regions, “mountain sites” were set to include at least one‐thirdopen mountain habitat and to be above a certain altitude, depending
on local conditions such as climate, latitude and historical land use.
These altitude thresholds for mountain sites were set to 400 m for
UK upland (and where the surveyed habitats were generally open),
1,100 m for the Giant Mountains and 1,200 m for all the remaining
southern mountains, respectively. The UK upland have a particularly
long history of anthropogenic deforestation and in combination with
high levels of extensive grazing and climatic exposure. Therefore,
open habitats resembling those of montane and alpine areas exist at
lower altitudes than would naturally occur (Smout, 2005; Thompson,
MacDonald, Marsden, & Galbraith, 1995). Also in the southern and
central European sites, open areas above the altitude limit are not
necessarily caused by the natural tree line, but areas also include
subalpine meadows that remain open due to grazing. The number of
study sites in each area is given in Table 1.
To define species which have significant populations in high‐alti-tude mountain habitats (so‐called mountain species), we used alti-
tude information from each larger mountain range area using data
from the United Kingdom (line transects, UK upland) and Switzerland
(territory mapping, the Alps) and Spain (line transects, Catalonian
Pyrenees). First, we calculated relative densities based on mountain
site‐specific species abundances and sampling effort (birds/km line
transect) in 100‐m‐altitude zones starting from the above‐mentioned
mountain thresholds of the regions. Second, based on altitude zone
densities, we calculated the mean altitudes of species for each
mountain region. In the United Kingdom, species whose mean alti-
tude were above 550 m (a.s.l.; more than half of the population
should be breeding above this altitude in mountain routes) and pre-
ferred open mountain habitats were included (Supporting Informa-
tion Table S2). We calculated mean altitudes separately for the
Swiss Alps and the Catalonian Pyrenees and used the mean of these
two values for both “Iberia” and “Alps.” The altitude threshold for
the species in these areas was above 1800 metres (Supporting Infor-
mation Table S3). In Fennoscandia, a set of 14 common mountain
species were already defined by Lehikoinen et al. (2014). However,
due to an increased monitoring effort in recent years, we could
include nine additional, less common, mountain species for this
region (Table 2).
We calculated species‐specific population trends for each of the
four defined mountain regions: Fennoscandia, UK upland, “Iberia”
and “Alps.” In addition, we pooled the counts from all regions to cal-
culate species trends for the whole area (further details are given
below). Trend analyses were conducted for species which had at
least five records per year in a given area (at the regional level, maxi-
mally 1 year with a sample size below five individuals was accepted).
When calculating the population trends for Europe, we also included
counts from mountain regions which had lower than five records
annually to maximize the total sample sizes. Mean annual sample
sizes are shown in Supporting Information Table S4.
Species were classified into mountain specialists or generalists,
based on their distribution areas in Europe. Species mainly restricted
to mountain areas and uncommon in the lowlands were classified as
mountain specialists whereas species which have substantial popula-
tions in the mountains but also commonly breed in lowlands were
classified as mountain generalists (Hagemeijer & Blair, 1997, see also
Scridel et al., 2018; Thompson, Kålås, & Byrkjedal, 2012; Table 2).
Furthermore, species were grouped into long distance (wintering in
tropical areas) and others (including both species wintering in the
Western Palearctic and residents) based on their distribution ranges
UK upland
Fennoscandia
F IGURE 1 A map showing the four European mountain regions,where the data was collected. The dots show the census locations(survey route) except in Italy where each dot represents one pointof a point count route
4 | LEHIKOINEN ET AL.
TABLE 2 The average annual population growth rates (trends) and traits of 44 mountain bird species in 11 European mountain areas, aswell as separate species trends for the “Alps”, Fennoscandia, “Iberia” and UK upland during 2002–2014. Traits include specialization(Sp = mountain specialists, G = generalists; classification based on distribution areas of Hagemeijer and Blair (1997)) and migratory behaviour(Ld = long‐distance migrant, Ot = other). Significant population change rates are in bold. “–” means that the species is not a typical mountainbird in the particular mountain region and NE means that species is a typical mountain species in the area, but there were too little dataavailable to calculate trends (see also Supporting Information Table S4)
Species (specialization) TraitsAll areas “Alps” Fennoscandia “Iberia” UKSlope ± SE Slope ± SE Slope ± SE Slope ± SE Slope ± SE
Clangula hyemalis Sp, Ot −0.033 ± 0.023 – −0.033 ± 0.023 – –
Buteo buteo G, Ot −0.006 ± 0.014 – – – −0.006 ± 0.014
Buteo lagopus G, Ot −0.041 ± 0.027 – −0.041 ± 0.027 – –
Falco tinnunculus G, Ot 0.008 ± 0.007 0.011 ± 0.008 ‐ −0.011 ± 0.021 –
Lagopus lagopus G, Ot −0.026 ± 0.006 – −0.095 ± 0.010 – 0.003 ± 0.007
Lagopus muta Sp, Ot −0.018 ± 0.008 0.013 ± 0.012 −0.047 ± 0.013 NE NE
Tetrao tetrix G, Ot 0.010 ± 0.027 0.035 ± 0.039 ‐ – NE
Alectoris graeca Sp, Ot 0.019 ± 0.021 0.019 ± 0.021 ‐ – –
Charadrius hiaticula G, Ot 0.050 ± 0.020 – 0.051 ± 0.021 – –
Charadrius morinellus Sp, Ot 0.012 ± 0.022 – 0.035 ± 0.024 – NE
Pluvialis apricaria G, Ot 0.013 ± 0.005 – 0.010 ± 0.005 – 0.022 ± 0.012
Calidris alpina G, Ot 0.005 ± 0.018 – 0.009 ± 0.021 – NE
Gallinago gallinago G, Ot −0.011 ± 0.012 – – – −0.011 ± 0.012
Tringa totanus G, Ot 0.033 ± 0.010 – 0.033 ± 0.010 – –
Phalaropus lobatus G, Ld −0.003 ± 0.030 – −0.003 ± 0.030 – –
Stercorarius longicaudus Sp, Ld 0.014 ± 0.017 – 0.014 ± 0.017 – –
Cuculus canorus G, Ld −0.053 ± 0.007 – −0.053 ± 0.007 – –
Alauda arvensis G, Ot −0.001 ± 0.003 0.016 ± 0.006 – −0.033 ± 0.008 0.004 ± 0.006
Hirundo rupestris Sp, Ot 0.001 ± 0.009 0.012 ± 0.011 – −0.017 ± 0.015 –
Anthus pratensis G, Ot −0.008 ± 0.003 NE −0.012 ± 0.005 NE −0.005 ± 0.004
Anthus spinoletta Sp, Ot −0.001 ± 0.003 0.000 ± 0.003 – −0.037 ± 0.013 –
Prunella collaris Sp, Ot 0.002 ± 0.007 0.002 ± 0.007 – NE –
Luscinia svecica G, Ld −0.001 ± 0.007 – −0.002 ± 0.008 – –
Phoenicurus ochruros G, Ot 0.008 ± 0.003 0.014 ± 0.003 – −0.025 ± 0.007 –
Phoenicurus phoenicurus G, Ld 0.014 ± 0.007 – 0.014 ± 0.007 – –
Saxicola rubetra G, Ld −0.030 ± 0.008 −0.029 ± 0.008 – −0.023 ± 0.049 –
Oenanthe oenanthe G, Ld 0.009 ± 0.003 0.026 ± 0.004 −0.005 ± 0.008 −0.013 ± 0.007 0.002 ± 0.008
Monticola saxatilis Sp, Ld −0.022 ± 0.013 −0.002 ± 0.017 – −0.059 ± 0.021 –
Turdus torquatus Sp, Ot 0.005 ± 0.004 0.001 ± 0.004 0.060 ± 0.025 0.000 ± 0.021 −0.006 ± 0.017
Turdus iliacus G, Ot −0.033 ± 0.006 – −0.033 ± 0.006 – –
Sylvia curruca G, Ld 0.011 ± 0.006 0.011 ± 0.006 – – –
Phylloscopus trochilus G, Ld −0.032 ± 0.003 – −0.032 ± 0.003 – –
Pyrrhocorax graculus Sp, Ot −0.015 ± 0.011 −0.002 ± 0.012 – −0.044 ± 0.025 –
Pyrrhocorax pyrrhocorax G, Ot 0.050 ± 0.012 NE – 0.053 ± 0.014 –
Corvus corone G, Ot −0.047 ± 0.014 – – – −0.047 ± 0.014
Corvus corax G, Ot 0.016 ± 0.013 – – – 0.016 ± 0.013
Montifringilla nivalis Sp, Ot 0.021 ± 0.010 0.021 ± 0.010 – NE –
Fringilla montifringilla G, Ot −0.025 ± 0.005 – −0.025 ± 0.005 – –
Serinus citrinella Sp, Ot −0.026 ± 0.013 −0.051 ± 0.031 – −0.023 ± 0.016 ‐
Carduelis cannabina G, Ot 0.015 ± 0.007 0.007 ± 0.008 – 0.040 ± 0.022 ‐
Carduelis flammea G, Ot −0.048 ± 0.005 −0.025 ± 0.007 −0.052 ± 0.007 – –
(Continues)
LEHIKOINEN ET AL. | 5
in winter (Cramp, Simmons, & Perrins, 1977–1994; Lehikoinen et al.,
2014).
2.3 | Weather data
We used European weather data (available at European Climate
Assessment & Dataset http://www.ecad.eu/download/ensembles/d
ownload.php in 0.25 degree grids across the continent) to calculate
changes in the temperature of the breeding season April–August.We tested rate of change in the mean temperature in each region in
the long term (1980–2014) and short term (1995–2014) using linear
regression. We first calculated region‐specific annual mean tempera-
tures from weather sites situated in the mountain region and then
conducted the linear regression. The locations from where the data
were extracted are shown in Supporting Information Figure S1.
2.4 | Statistical analyses
Log‐linear population trends and annual indices were calculated for
each species separately using the software TRIM (Pannekoek & Van
Strien, 2005). TRIM is a commonly used tool in bird monitoring in
Europe that accounts for overdispersion and serial correlation and
interpolates missing observations using a Poisson general log‐linearmodel (European Bird Census Council, 2018). TRIM produces annual
growth rate as well as annual abundance indices, including their
standard errors. Long‐term annual growth rates and annual abun-
dance indices were calculated for Europe using aggregated data from
all regions and separately for each of the four major mountain
regions. We compared the change in the overall mountain bird indi-
cator to the corresponding magnitude of change in European (a)
common bird, (b) farmland and (c) forest bird indicators during
2002–2014 provided by European Bird Census Council (2018).
The calculation of the indicators was done using a new statistical
tool, which has not been used earlier in continental analyses. We
combined annual population indices of species as multi‐species indi-
cators using the R‐package tool (Soldaat, Pannekoek, Verweij, van
Turnhout, & van Strien, 2017). The package calculates annual multi‐species indicator values and their standard errors as well as a long‐term change of the indicator using Monte Carlo simulation method
and the species‐specific indices and their standard errors provided
by TRIM. We used TREND_DIFF‐function of the package to test
whether the indicators differed from each other (specialist vs. gener-
alists, or regional indicators).
Spatial differences in sampling network could lead into a situa-
tion where trends are more driven by areas where number of census
sites are dense compared to areas where the network is sparse. We
therefore, per each contributing country, weighted the trend analy-
ses by the spatial coverage of the national network. As weight, we
used the country‐specific mountain region area divided by the num-
ber of census sites (average area per census sites: larger value mean
lower density of census sites). Thus, census sites in countries with
proportionally fewer routes in mountain areas weighed more in the
analyses. France contributed to data of two regions (“Iberia” and
“Alps”), and thus, the weights were calculated separately for these
regions. The mountain area was measured using Corine land cover
data (Copernicus Land Monitoring Service, 2016), where mountain
habitats were (a) natural grasslands, (b) moors and heathlands, (c)
transitional woodland shrubs, (d) bare rock, (e) sparsely vegetated
areas, (f) glaciers and perpetual snow and (g) peat bogs, which were
above certain region‐specific altitude (see Supporting Information
Table S5). Here, we have used the data of the year 2012 only. We
believe that this represents the general situation in each country,
because these habitat types unlikely show large‐scale changes during
the relatively short study period.
Last, we analysed a set of factors that potentially could explain
the regional population trends of species provided by TRIM analyses
in the four major mountain areas during 2002–2014, using GLMM
(functions lmer and lmerTest in R). Regional long‐term population
trends were tested against migratory behaviour (long‐distancemigrants or other, the latter including residents, which are rare
among mountain birds), specialization (mountain specialists or gener-
alists) and short‐term temperature change in each region (“Alps,”
Fennoscandia, “Iberia” and the United Kingdom; Table 3). Species
was a random factor in the model to account for some species hav-
ing data from several mountain regions whereas some only have
data from one of them. We took phylogeny into account in the anal-
yses since species with the same ancestors may have more similar
responses. We did this by first using various phylogenic structures
(order, family and genus based on del Hoyo, Collar, Christie, Elliot,
and Fishpool (2014) and del Hoyo et al. (2016), altogether eight
combinations, see Supporting Information Table S6) in the random
part of the full model. We ranked these models based on AICc
(Burnham & Anderson, 2004). Second, we used the best phylogenic
structure in the final analyses, where we constructed 12 model com-
binations, and where the full model included the two‐way interac-
tions temperature*migration and temperature*specialization. The
inclusion of an interaction between temperature and migration was
based on the hypothesis that species that spend most of the time in
the mountain areas (short‐distance migrants and residents) may face
the largest declines in areas where the temperature increase has
TABLE 2 (Continued)
Species (specialization) TraitsAll areas “Alps” Fennoscandia “Iberia” UKSlope ± SE Slope ± SE Slope ± SE Slope ± SE Slope ± SE
Calcarius lapponica Sp, Ot −0.026 ± 0.008 – −0.026 ± 0.008 – –
Plectrophenax nivalis Sp, Ot −0.041 ± 0.014 – −0.042 ± 0.014 – NE
Emberiza cia Sp, Ot −0.031 ± 0.006 −0.024 ± 0.012 – −0.033 ± 0.008 –
6 | LEHIKOINEN ET AL.
been highest. The interaction between temperature and specializa-
tion relates to the hypothesis that specialists would be declining
fastest in the area with high temperature increase. The model combi-
nations are shown in Table 3. These 12 models were ranked based
on AIC corrected for small sample sizes (Burnham & Anderson,
2004). Finally, we took the uncertainty of the population trends into
account in the analyses using the reciprocal of the standard errors of
the trends as weights. We used R (version 3.4.1) in all the analyses
(R Core Development Team, 2017).
3 | RESULTS
Because the results of the weighted analyses according to the
national area per census sites ratio were almost identical to the non‐weighted analyses (Supporting Information Table S6), we decided to
show only the unweighted results in the main results section
(Table 2).
The European mountain bird indicator showed a significant nega-
tive decline during 2002–2014 (44 species; −0.61%/year, 95% CI
−1.14 to −0.08, overall decline c. −7%; Figure 2a). The European
mountain specialist indicator also declined significantly (n = 16 spe-
cies, −0.88%/year, 95% CI −1.66 to −0.10, overall decline c. −10%).
The mountain generalist slope was also negative (n = 28 species,
−0.46%/year), but not significantly so (95% CI −1.06 to 0.17; Fig-
ure 2b). The slopes of specialists and the generalists did not differ
from each other (trend difference = 0.0040, se = 0.0051, p > 0.05,
see also Table 3). Among the specialists, five out of 16 species
showed negative and one showed positive trends. Among the gener-
alists, nine out of 28 species declined and seven increased (Table 2).
Despite the fact that many mountain bird species have a wide distri-
bution in Europe, it is important to note that only for two out of 44
species (northern wheatear and ring ouzel) were there enough data
to calculate trends in all four mountain areas. In addition, for about
half of the species, population trends were only calculated for one
of the four regions, because the species were too rare in other
regions (Table 2).
The indicator of “Alps” showed no significant trends during
2002–2014 (n = 20 species, +0.29%/year, 95% CI −0.59 to 1.17,
Figure 3a). Four species showed positive, and three species showed
negative trends during 2002–2014 (Table 2). The Fennoscandian and
“Iberian” indicators showed significant negative trends during 2002–2014 (Fennoscandia, n = 23 species, −1.20%/year, 95% CI −2.04 to
−0.36, overall decline −13%; “Iberia”, n = 14 species, −1.94%, 95%
CI −3.61 to −0.27, overall decline −21%; Figure 3b–c). In
Fennoscandia and “Iberia,” respectively, 10 and five species showed
negative, and three and one showed positive trends (Table 2). The
indicator of UK upland showed no significant trend during 2002–2014 (n = 10 species, −0.29%/year, 95% CI −1.13 to 0.55, Figure 3d).
In UK upland, one species declined (carrion crow) and none
increased in 2002–2014 (Table 2). According to bootstrapping simu-
lations, the slopes of Fennoscandian and “Iberian” indicators differed
significantly from slopes in the “Alps” (trend difference between
“Alps” and Fennoscandia 0.015 ± 0.006 se, p < 0.05, trend differ-
ence between “Alps” and Iberia 0.022 ± 0.010 se, p < 0.05). Slopes
of the other regions did not differ from each other (all p > 0.05).
The species only was the best random structure compared to
more complicated phylogenic structures (Supporting Information
Table S7), and thus, species only was used in the latter analyses. The
best model explaining the regional population trends of species dur-
ing 2002–2014 was the null model. Although two other more com-
plex models were within two AIC units, additional variables of those
models can be considered as uninformative parameters (sensu
Arnold, 2010). Thus, this modelling approach was not able to find
that region, specialization or migratory behaviour was linked with
the regional population trends (Table 3). The intercept of the null
model was significantly below zero (−0.0072 ± 0.0035, t = 2.0,
p < 0.05), suggesting in general negative regional population trends
during this particular period.
Annual temperatures during the breeding season (April–August)increased significantly in all four regions in the long term (rate of
increase 0.81–1.55°C during 1980–2014; Table 4). During the last
20 years (1995–2014), the temperature increase was only significant
in Fennoscandia (Table 4).
4 | DISCUSSION
We set out to test three hypotheses regarding the recent population
trends in European mountain birds. We got unequivocal support for
the first hypothesis regarding a negative trend of European mountain
bird populations since we found that the indicator has an overall
decline of −7% during 2002–2014 (−0.61%/year). Fennoscandian
and “Iberian” mountain bird indicators declined significantly and dif-
fered from the slope of the corresponding indicator in the “Alps”.
Based on European common bird monitoring, the magnitude of the
decline is the same as all common birds in Europe during the same
TABLE 3 AICc differences, AIC weights (w) and evidence ratios(ER) of models explaining regional population trends of mountainbirds during 2002–2014. Spe is specialization (mountain specialist orgeneralist), Mig is migratory behaviour (short‐ or long‐distancemigrant) and Mt is mountain region
Model ΔAICc w ER
Intercept only 0.00 0.276 1.0
Temp 0.96 0.171 1.6
Spe 1.53 0.128 2.2
Mig 2.05 0.099 2.8
Spe + Temp 2.35 0.085 3.2
Spe + Temp + Spe*Temp 3.13 0.057 4.8
Mig + Temp 3.22 0.055 5.0
Mig + Spe 3.43 0.050 5.5
Mig + Spe + Temp 4.53 0.029 9.5
Mig + Spe + Temp + Spe*Temp 5.45 0.018 15.3
Mig + Temp + Mig*Temp 5.46 0.018 15.3
Mig + Spe + Temp + Mig*Temp 6.87 0.009 30.7
LEHIKOINEN ET AL. | 7
study period. More specifically, the trends of bird indicators in two
important habitats, farmland and forests, were −13% and −1%, dur-
ing the study same period, respectively (European Bird Census
Council, 2018). Thus, in general mountain birds are doing less bad
than farmland birds, but clearly worse than forest birds in Europe.
The severe declines of farmland birds are mainly driven by intensifi-
cation of agriculture rather than climate change (Butler, Boccacio,
Gregory, Voříšek, & Norris, 2010; Eglington & Pearce‐Higgins, 2012;
Jørgensen et al., 2016). However, in case of mountain birds, climate
change can have a larger impact as the climatic niche of especially
mountain specialists is shrinking, highlighted by the relatively fast
declines of mountain species.
As far as our second hypothesis is concerned, that the decline
would be stronger in mountain specialists than in mountain
generalists, the outcomes of our tests are less straightforward to
interpret. Numerically, the decline was indeed larger among the spe-
cialists (−0.88%/year vs. −0.46%/year). However, the two slopes
were not statistically different from each other, nor is the generalist
slope statistically significant in itself. We believe that the nonsignifi-
cant difference between these two groups is at least partly caused
by small sample sizes, which increase uncertainty in the trend esti-
mates and reduce statistical power. The topic should be re‐evaluatedin the future with longer time series. In general, we should be more
worried about mountain specialists, since this group of species
showed already significant population declines.
We got no support for our third main hypothesis that long‐dis-tance migrant mountain birds have fared worse than resident and
short‐distance migrant mountain birds, finding no significant differ-
ences between migratory groups on the regional level. Therefore,
the diminishing mountain bird populations are not only driven by
general declines of long‐distance migrants (e.g., Sanderson et al.,
2006; Vickery et al., 2014), but also species wintering in Europe are
contributing to the decline in mountain birds. This could indicate
that mountain species have also problems in their breeding areas
(Lehikoinen et al., 2014). More work needs to be done to under-
stand, what are the valid traits to evaluate the vulnerability of moun-
tain species in the face of climate change (see also MacLean &
Beissinger, 2017).
60 70 80 90
100 110 120 130
2002 2004 2006 2008 2010 2012 2014 60 70 80 90
100 110 120 130
2002 2004 2006 2008 2010 2012 2014 60
70
80
90
100
110
120
130
2002 2004 2006 2008 2010 2012 2014
Generalists
Specialists
(b)
Year
Inde
x (a)
F IGURE 2 (a) The mountain birdindicator for Europe and (b) the separateindicators for specialists and generalists,during 2002–2014. Calculated mean of theindices and their 95% CIs are given
60
80
100
120
140
2002 2004 2006 2008 2010 2012 2014 60
80
100
120
140
2002 2004 2006 2008 2010 2012 2014
UK
Fennoscandia
(c)
Year
Inde
x
(a) (b)
(d)
“Alps”
“Iberia”
F IGURE 3 Regional mountain birdindicators during 2002–2014 from (a)“Alps”, (b) Fennoscandia, (c) “Iberia” and (d)United Kingdom. Calculated mean of theindices and their 95% CIs are given
TABLE 4 Annual changes in temperature (in °C from April toAugust) in four mountain regions in Europe during 1980–2014 and1995–2014. Significant temperature changes are marked in bold
Mountain area 1980–2014 1995–2014
“Alps” 0.045 ± 0.012 0.016 ± 0.026
Fennoscandia 0.035 ± 0.012 0.067 ± 0.031
“Iberia” 0.037 ± 0.010 0.013 ± 0.026
UK upland 0.024 ± 0.008 0.007 ± 0.019
8 | LEHIKOINEN ET AL.
The reason why there seem to be no universal patterns
explaining species‐specific variation in responses to climate change
could be that regional circumstances, such as land use practices,
differ between areas. In one area, impacts of climate change may
be more important than changes in land use and vice versa. Agro‐pastoral land use practices have become less intense or have
been abandoned completely allowing forest cover to increase
again, especially in the low‐altitude mountains of the southern
mountain regions (“Alps” and “Iberia”; Brambilla et al., 2010; Her-
rando et al., 2016; Maggini et al., 2014). Interactions with agricul-
tural abandonment and forest expansion can be complex and
offer both threats and opportunities depending on the ecological
requirements of species and assemblages involved (Calladine,
Bielinski, & Shaw, 2013; Gillings, Fuller, & Henderson, 1998; Her-
rando et al., 2016).
The April–August temperatures have increased substantially in
recent decades in all four mountain areas. Although the tempera-
ture increase has been significant only in Fennoscandia over the
last two decades, the temperatures are nowadays above the long‐term mean in all regions (Lehikoinen et al., 2014). Climate change
may affect bird populations in a different manner depending on
the region (Sæther & Engen, 2010). Furthermore, temperatures are
expected to rise faster in higher northern latitude mountains than
in mountains located in temperate and tropical zones, and the rate
of warming in mountain systems can be two to three times higher
than that recorded during the 20th century (Nogués‐Bravo et al.,
2007). These can cause considerable effects on biodiversity even
though the direct impacts can be difficult to measure (Araújo,
Errea, & Martinez‐Rica, 2007). Although we could not link the
population dynamics with the observed climate change, the
observed declines are in line with the population predictions in
relation to climate change (Huntley et al., 2007). Human‐inducedland use changes are not as extensive in Fennoscandian mountains
(Lehikoinen et al., 2014) compared to “Iberia” (Herrando et al.,
2016), and several Fennoscandian studies have revealed changes in
plant community due to climate change (Kullman & Öberg, 2009;
Michelsen, Syverhuset, Pedersen, & Holten, 2011; Vuorinen et al.,
2017). One should also keep in mind that especially in Fennoscan-
dia some mountain species are nomadic to some extent (Lind-
ström, 1987) and both plant and animal communities are strongly
influenced by multi‐annual cyclic fluctuation of small rodents (Han-
ski, Hansson, & Henttonen, 1991; Turchin, Oksanen, Ekerholm,
Oksanen, & Henttonen, 2000). Even animal species that are not
using rodents in their diet are influenced by the cycles due to
predator–prey interactions (Lehikoinen et al., 2016). Despite these
kinds of fluctuations, we were able to detect a negative long‐termtrend in Fennoscandia.
We must stress that the methods of the monitoring schemes and
their intensity showed spatial variation within the overall study area.
However, we do not believe that this has biased the analysis. First,
the magnitude of the trend should be comparable independently of
whether it is based on point count, line transect or territory mapping
(Gregory et al., 2005). Second, we tried to compensate for the
potential biases in the sampling by using country‐specific weights.
The use of weights did not influence the main results. We believe
that there are two reasons why our weighting did not influence the
population trends: (a) Many of the species data are only available
from one of the study regions and thus weighting between regions
have no importance; and (b) population trends of nearby countries
are similar. As the monitoring schemes have improved in many coun-
tries in recent years including systematic sampling, future analyses of
monitoring data will be even more reliable due to increased sample
sizes.
Modelling work on the future effects of climate and land use
change have suggested that species‐specific conservation measures
aiming at improving habitat to counteract the negative influence of
climate change can only deliver minor improvements of the future
fate of mountain birds (Braunisch et al., 2014). Even if high moun-
tains may provide refuges for threatened mountain species currently
populating lower altitudes, in the long term, climate change can be
expected to have a strong impact on alpine species (Freeman, Scho-
ler, Ruiz‐Gutierrez, & Fitzpatrick, 2018). Alpine habitats are expected
to be reduced and become more fragmented and isolated due to rise
of the tree line where species have increasing limited dispersal possi-
bilities. Our findings also emphasize that local studies are needed to
understand the mechanisms and drivers of the population changes
of individual species and species communities in mountains including
information about species habitat selection and changes in the
amount of preferred habitat. Despite international actions to halt cli-
mate change, climate will change in the near future (EEA, 2012). To
mitigate the potential impacts of climate change, it is important to
take measures that can improve connectivity between suitable
mountain habitats and to minimize the effects of other threats such
as non‐sustainable tourism and afforestation of grasslands (Lloret,
2017).
Last, to understand the big picture on the continental and global
scale we also need to continue existing monitoring work in the
mountain areas and expand both the taxonomic and spatial coverage
of monitoring schemes. Monitoring should preferably be based on
systematic sampling design with a reasonable number of study sites
covered on annual basis. One reason why we did not observe signifi-
cant differences in trends between specialization groups could be
the still relatively small sample sizes and thus larger uncertainties in
our trend estimates. Nevertheless, our European mountain bird indi-
cator and regional indicators provide an important tool to measure
and monitor the changes in mountain biodiversity with regular
updates in the future and the spatial coverage of the indicator can
easily be expanded when suitable monitoring data become available.
Given that climate and land use changes in the uplands are likely to
manifest themselves into the loss of open mountain habitats and
expansion of shrubland/forest, we suggest that future work should
also look at mechanistic reasons behind the declines. More and
important information may come from comparing potential differ-
ences in trends between mountain and lowland population of the
mountain generalists, where the land use pressures can differ
between the areas.
LEHIKOINEN ET AL. | 9
ACKNOWLEDGEMENTS
We wish to thank all the volunteers participating censuses. The
Finnish common bird monitoring has been supported by the Finnish
Ministry of the Environment. The surveys of the fixed routes in
Sweden were supported by grants from the Swedish Environmental
Protection Agency and carried out in collaboration with all 21
County Administrative Boards of Sweden. The bird surveys are car-
ried out within the framework of the Centre for Animal Movement
Research and the strategic research environment Biodiversity and
Ecosystem Services in a Changing Climate (BECC). The Norwegian
Terrestrial Bird Monitoring programme is financed by the Climate
and Environment Ministry and the Norwegian Environment Agency.
The Catalan Common Bird Survey is supported by the Catalan
Government. The Italian Common Breeding Birds Monitoring Pro-
gram has been partly funded by local governments or private orga-
nizations, with a great effort in fieldwork produced by volunteers
till 2008; since 2009 the Program has been almost completely
funded by the Ministry of Agricultural, Food and Forestry Policies
and by the Rural Network, with the exception of few Regions
where local governments went on funding monitoring activities. AL
received financial support from the Academy of Finland (275606).
In the United Kingdom, the BTO/JNCC/RSPB Breeding Bird Survey
is a partnership jointly funded by the BTO, RSPB and JNCC, with
fieldwork undertaken by volunteers and the additional mountain
transect surveys were undertaken by BTO supported by the Scot-
tish Government, Scottish Natural Heritage, the National Trust for
Scotland, the Scottish Ornithologists' Club, the Scottish Moun-
taineering Trust and the AEB Trust.
ORCID
Aleksi Lehikoinen https://orcid.org/0000-0002-1989-277X
Åke Lindström https://orcid.org/0000-0002-5597-6209
REFERENCES
Araújo, D., Errea, M. B., & Martinez‐Rica, M. P. (2007). Exposure of global
mountain systems to climate warming during the 21st Century. Glo-
bal Environmental Change, 17, 420–428. https://doi.org/10.1016/
j.gloenvcha.2006.11.007
Arlettaz, R., Patthey, P., Baltic, M., Leu, T., Schaub, M., Palme, R., &
Jenni‐Eiermann, S. (2007). Spreading free‐riding snow sports repre-
sent a novel serious threat for wildlife. Proceedings of the Royal Soci-
ety B‐Biological Sciences, 274, 1219–1224. https://doi.org/10.1098/
rspb.2006.0434
Arnold, T. W. (2010). Uninformative parameters and model selection
using Akaike's Information Criterion. Journal of Wildlife Management,
74, 1175–1178. https://doi.org/10.1111/j.1937-2817.2010.tb01236.x
Baillie, J. E. M., Hilton‐Taylor, C., & Stuart, S. N. (Eds.) (2004). IUCN red
list of threatened species. A global species assessment (xxiv + pp.
191). Gland, Switzerland: IUCN.
Brambilla, M., Casale, F., Bergero, V., Bogliani, G., Crovetto, G. M., Falco,
R., … Negri, I. (2010). Glorious past, uncertain present, bad future?
Assessing effects of land‐use changes on habitat suitability for a
threatened farmland bird species. Biological Conservation, 143, 2770–2778. https://doi.org/10.1016/j.biocon.2010.07.025
Braunisch, V., Coppes, J., Arlettaz, R., Suchant, R., Zellweger, F., & Boll-
mann, K. (2014). Temperate mountain forest biodiversity under cli-
mate change: Compensating negative effects by increasing structural
complexity. PLoS ONE, 9(5), e97718. https://doi.org/10.1371/journal.
pone.0097718
Breed, G. A., Stichter, S., & Crone, E. (2013). Climate‐driven changes in
northeastern US butterfly communities. Nature Climate Change, 3,
142–145. https://doi.org/10.1038/nclimate1663
Britton, A. J., & Fisher, J. M. (2007). Interactive effects of nitrogen depo-
sition, fire and grazing on diversity and composition of low‐alpineprostrate Calluna vulgaris heathland. Journal of Applied Ecology, 44,
125–135. https://doi.org/10.1111/j.1365-2664.2006.01251.xBurnham, K. P., & Anderson, D. R. (2004). Multimodel inference – Under-
standing AIC and BIC in model selection. Sociological Methods &
Research, 33, 261–304. https://doi.org/10.1177/0049124104268644Butchart, S. H. M., Stattersfield, A. J., Baillie, J., Bennun, L. A., Stuart, S.
N., Akcakaya, H. R., … Mace, G. M. (2005). Using Red List Indices to
measure progress towards the 2010 target and beyond. Philosophical
Transactions of the Royal Society B: Biological Sciences, 360, 255–268.Butchart, S. H. M., Walpole, M., Collen, B., van Strien, A., Scharlemann, J.
P. W., Almond, R. E. A., … Watson, R. (2010). Global biodiversity:
Indicators of recent declines. Science, 328, 1164–1168. https://doi.org/10.1126/science.1187512
Butler, S. J., Boccaccio, L., Gregory, R. D., Voříšek, P., & Norris, K. (2010).
Quantifying the impact of land‐use change to European farmland bird
populations. Agriculture, Ecosystems & Environment, 137, 348–357.https://doi.org/10.1016/j.agee.2010.03.005
Calladine, J., Bielinski, A., & Shaw, G. (2013). Effect on bird abundance
and species richness of edge restructuring to include shrubs at the
interface between conifer plantations and open moorland. Bird Study,
60, 345–356. https://doi.org/10.1080/00063657.2013.797953Chamberlain, D., Arlettaz, R. L., Caprio, E., Maggini, R., Pedrini, P.,
Rolando, A., & Zbinden, N. (2012). The altitudinal frontier in avian cli-
mate impact research. Ibis, 154, 205–209. https://doi.org/10.1111/j.1474-919X.2011.01196.x
Chamberlain, D. E., & Fuller, R. J. (2001). Contrasting patterns of change
in the distribution and abundance of farmland birds in relation to
farming system of lowland Britain. Global Ecology and Biogeography,
10, 399–409. https://doi.org/10.1046/j.1466-822x.2001.00236.xChen, I. C., Hill, J. K., Ohlemüller, R., Roy, D. B., & Thomas, C. D. (2011).
Rapid range shifts of species associated with high levels of climate
warming. Science, 333, 1024–1026. https://doi.org/10.1126/science.1206432
Collen, B., Loh, J., Whitemee, S., McRae, L., Amin, R., & Baillie, J. E. M.
(2009). Monitoring change in vertebrate abundance: The Living Pla-
net Index. Conservation Biology, 23, 317–327. https://doi.org/10.
1111/j.1523-1739.2008.01117.x
Copernicus Land Monitoring Service. (2016). Corine land cover (CLC)
2012, version 18.5.1. Retrieved from http://land.copernicus.eu/pan-
european/corine-land-cover/clc-2012
Cramp, S., Simmons, K. E. L., & Perrins, C. M. (1977–1994). Handbook of
the birds of Europe, Middle East and North America: Birds of the Wes-
tern Palaearctic. Oxford, UK: Oxford University Press.
Davey, C. M., Chamberlain, D. E., Newson, S. E., Noble, D. G., & John-
ston, A. (2012). Rise of the generalists: Evidence for climate driven
homogenization in avian communities. Global Ecology and Biogeogra-
phy, 5, 568–578. https://doi.org/10.1111/j.1466-8238.2011.00693.xDavey, C. M., Devictor, V., Jonzén, N., Lindström, Å., & Smith, H. G.
(2013). Impact of climate change on communities: Revealing species’contributions. Journal of Animal Ecology, 82, 551–561. https://doi.
org/10.1111/1365-2656.12035
del Hoyo, J., Collar, N. J., Christie, D. A., Elliott, A., & Fishpool, L. D. C.
(2014). HBW and Birdlife International illustrated checklist of the birds
10 | LEHIKOINEN ET AL.
of the world. Non‐passerines (Vol. 1). Barcelona, Spain and Cambridge,
UK: Lynx Edicions and BirdLife International.
del Hoyo, J., Collar, N. J., Christie, D. A., Elliott, A., Fishpool, L. D. C.,
Boesman, P., & Kirwan, G. M. (2016). HBW and BirdLife International
illustrated checklist of the birds of the world. Passerines (Vol. 2). Barce-
lona, Spain and Cambridge, UK: Lynx Edicions and BirdLife Interna-
tional.
Devictor, V., van Swaay, C., Brereton, T., Brotons, L., Chamberlain, D.,
Heliölä, J., … Jiguet, F. (2012). Differences in the climatic debts of
birds and butterflies at a continental scale. Nature Climate Change, 2,
121–124. https://doi.org/10.1038/nclimate1347
Eglington, S. M., & Pearce‐Higgins, J. W. (2012). Disentangling the rela-
tive importance of changes in climate and land‐use intensity in driv-
ing recent bird population trends. PLoS ONE, 7, e30407. https://doi.
org/10.1371/journal.pone.0030407
European Bird Census Council. (2018). Pan-European Common Bird
Monitoring Scheme (PECBMS). Available at: www.ebcc.info. Cited on
1.12.2017.
European Environment Agency (EEA). (2012). Climate change, impacts and
vulnerability in Europe 2012, an indicator based report. EEA Report No.
12/2012 (p. 304). Copenhagen, Denmark: European Environment
Agency (EEA).
Fahrig, L. (2003). Effects of habitat fragmentation on biodiversity. Annual
Review in Ecology, Evolution and Systematics, 34, 487–515. https://doi.org/10.1146/annurev.ecolsys.34.011802.132419
Flousek, J., Telenský, T., Hanzelka, J., & Reif, J. (2015). Population
trends of Central European montane birds provide evidence for
adverse impacts of climate change on high‐altitude species. PLoS
ONE, 10(10), e0139465. https://doi.org/10.1371/journal.pone.
0139465
Freeman, B. G., Scholer, M. N., Ruiz‐Gutierrez, V., & Fitzpatrick, J. W.
(2018). Climate change causes upslope shifts and mountaintop extir-
pations in a tropical bird community. Proceedings of the National
Academy of Science of the United States of America, 115(47), 11982–11987. https://doi.org/10.1073/pnas.1804224115
Gillings, S., Fuller, R. J., & Henderson, C. B. (1998). Avian community
composition and patterns of bird distribution within birch‐heathmosaics in north‐east Scotland. Ornis Fennica, 75, 27–37.
Gonzalez, P., Neilson, R. P., Lenihan, J. M., & Drapek, R. J. (2010). Global
patterns in the vulnerability of ecosystems to vegetation shifts due
to climate change. Global Ecology and Biogeography, 19, 755–768.https://doi.org/10.1111/j.1466-8238.2010.00558.x
Gough, L. A., Sverdrup‐Thygeson, A., Milberg, P., Pilskog, H. E., Jansson,
N., Jonsell, M., & Birkemoe, T. (2015). Specialists in ancients trees are
more affected by climate than generalists. Ecology and Evolution, 5,
5632–5641. https://doi.org/10.1002/ece3.1799Gregory, R. D., van Strien, A., Voříšek, P., Meyling, A. W. G., Noble, D.
G., Foppen, R. P. B., & Gibbons, D. W. (2005). Developing indicators
for European birds. Philosophical Transactions of the Royal Society B:
Biological Sciences, 360, 269–288.Gregory, R. D., Voříšek, P., Noble, D. G., Van Strien, A., Klvanová, A.,
Eaton, M., … Burfield, I. J. (2008). The generation and use of bird
population indicators in Europe. Bird Conservation International, 18,
S223–S244. https://doi.org/10.1017/S0959270908000312Hagemeijer, W. J. M., & Blair, M. J. (1997). The EBCC atlas of European
breeding birds: Their distribution and abundance. London, UK: T. & A.
D. Poyser.
Hanski, I., Hansson, L., & Henttonen, H. (1991). Specialist predators, gen-
eralist predators and the microtine rodent cycle. Journal of Animal
Ecology, 60, 353–367. https://doi.org/10.2307/5465Herrando, S., Brotons, L., Anton, M., Páramo, F., Villero, D., Titeux, N., …
Stefanescu, C. (2016). Assessing impacts of land abandonment on
Mediterranean biodiversity using indicators based on bird and butter-
fly monitoring data. Environmental Conservation, 43, 69–78. https://doi.org/10.1017/S0376892915000260
Huntley, B., Green, R. E., Collingham, Y. C., & Willis, S. G. (2007). A cli-
matic atlas of European breeding birds. Durham Univ, Barcelona: The
RSPB and Lynx Edicions.
Ims, R. A., & Henden, J.‐A. (2012). Collapse of an arctic bird community
resulting from ungulate‐induced loss of erect shrubs. Biological Con-
servation, 149, 2–5. https://doi.org/10.1016/j.biocon.2012.02.008Jacob, D., Petersen, J., Eggert, B., Alias, A., Christensen, O. B., Bouwer, L.
M., … Yiou, P. (2014). EURO‐CORDEX: New high‐resolution climate
change projections for European impact research. Regional Environ-
mental Change, 14, 563–578. https://doi.org/10.1007/s10113-013-
0499-2
Jørgensen, P. S., Böhning‐Gaese, K., Thorup, K., Tøttrup, A. P., Chylarecki,P., Jiguet, F., … Rahbek, C. (2016). Continent‐scale global change
attribution in European birds – Combining annual and decadal time
scales. Global Change Biology, 22, 530–543. https://doi.org/10.1111/gcb.13097
Kullman, L., & Öberg, L. (2009). Post‐Little Ice Age tree line rise and cli-
mate warming in the Swedish Scandes: A landscape ecological per-
spective. Journal of Ecology, 97, 415–429. https://doi.org/10.1111/j.1365-2745.2009.01488.x
Laaksonen, T. K., & Lehikoinen, A. (2013). Population trends in boreal
birds: Continuing declines in long‐distance migrants, agricultural and
northern species. Biological Conservation, 168, 99–107. https://doi.
org/10.1016/j.biocon.2013.09.007
Lehikoinen, A., Green, M., Husby, M., Kålås, J. A., & Lindström, Å. (2014).Common montane birds are declining in northern Europe. Journal of
Avian Biology, 45, 3–14. https://doi.org/10.1111/j.1600-048X.2013.
00177.x
Lehikoinen, A., Fraixedas, S., Burgas, D., Eriksson, H., Henttonen, H.,
Laakkonen, H., … Välimäki, K. (2016). The impact of weather and the
phase of the rodent cycle on breeding population of waterbirds in
Finnish Lapland. Ornis Fennica, 93, 31–46.Lindström, Å. (1987). Breeding nomadism and site tenacity in the Bram-
bling Fringilla montifringilla. Ornis Fennica, 64, 50–56.Lindström, Å., Green, M., Paulson, G., Smith, H. G., & Devictor, V. (2013).
Rapid changes in bird community composition at multiple spatial
scales in response to recent climate change. Ecography, 36, 313–322.https://doi.org/10.1111/j.1600-0587.2012.07799.x
Lloret, F. (2017). Trade‐offs in high mountain conservation. In J. Catalan,
J. M. Ninot, & M. M. Aniz (Eds.), High mountain conservation in a
changing world (pp. 37–59). Cham, Switzerland: Springer (SpringerLink
Bücher, 62).
MacLean, S., & Beissinger, S. R. (2017). Species’ traits as predictors of
range shifts under contemporary climate change: A review and meta‐analysis. Global Change Biology, 23, 4094–4105. https://doi.org/10.
1111/gcb.13736
Maggini, R., Lehmann, A., Kéry, M., Schmid, H., Beniston, M., Jenni, L., &
Zbinden, N. (2011). Are Swiss birds tracking climate change? Detect-
ing elevational shifts using response curve shapes. Ecological Model-
ling, 222, 21–32. https://doi.org/10.1016/j.ecolmodel.2010.09.010
Maggini, R., Lehmann, A., Zbinden, N., Zimmerman, N. E., Bolliger, J.,
Schröder, B., … Jenni, L. (2014). Assessing species vulnerability to cli-
mate and land use change. The case of the Swiss breeding birds.
Diversity Distribution, 20, 708–719. https://doi.org/10.1111/ddi.
12207
Michelsen, O., Syverhuset, A. O., Pedersen, B., & Holten, J. I. (2011). The
impact of climate change on recent vegetation changes on Dovrefjell,
Norway. Diversity, 3, 91–111. https://doi.org/10.3390/d3010091Nogués‐Bravo, D., Araújo, M., Errea, M. P., & Martinez‐Rica, J. P. (2007).
Exposure of global mountain systems to climate warming during the
21st Century. Global Environmental Change, 17, 420–428. https://doi.org/10.1016/j.gloenvcha.2006.11.007
Normander, B., Levin, G., Auvinen, A. P., Bratli, H., Stabbetorp, O., Hed-
blom, M., … Gundmundsson, G. A. (2012). Indicator framework for
measuring quantity and quality of biodiversity – Exemplified in the
LEHIKOINEN ET AL. | 11
Nordic countries. Ecological Indicators, 13, 104–116. https://doi.org/10.1016/j.ecolind.2011.05.017
Pannekoek, J., & van Strien, A. (2005). TRIM 3 manual (TRends and Indices
for Monitoring data). Voorburg, the Netherlands: Statistics Nether-
lands.
R Core Development Team (2017). R: A language and environment for sta-
tistical computing, version 3.4.1. Vienna, Austria: R Foundation for Sta-
tistical Computing. Retrieved from https://www.r-project.org/
Rodriguez‐Rodriguez, D., Bomhard, B., Butchart, S. H. M., & Foster, M. N.
(2011). Progress towards international targets for protected areas
coverage in mountains: A multi‐scale assessment. Biological Conserva-
tion, 144, 2978–2983. https://doi.org/10.1016/j.biocon.2011.08.023Root, T. L., Price, J. T., Hall, K. R., Schneider, S. H., Rosenzweig, C., &
Pounds, J. A. (2003). Fingerprints of global warming on wild animals
and plants. Nature, 421, 57–60. https://doi.org/10.1038/nature01333Sæther, B.‐E., & Engen, S. (2010). Population consequences of climate
change. In A. P. Møller, W. Fiedler, & P. Berthold (Eds.), Effects of cli-
mate change on birds. Oxford, UK: Oxford University Press.
Sanderson, F. J., Donald, P. F., Pain, D. J., Burfield, I. J., & van Bommel, F.
P. J. (2006). Long‐term population declines in Afro‐Palearctic migrant
birds. Biological Conservation, 131, 93–105. https://doi.org/10.1016/j.biocon.2006.02.008
Scridel, D., Brambilla, M., Martin, K., Lehikoinen, A., Iemma, A., Matteo,
A., … Chamberlain, D. (2018). A review and meta‐analysis of the
effects of climate change on Holarctic mountain and upland bird pop-
ulations. Ibis, 160, 489–515. https://doi.org/10.1111/ibi.12585Sekercioglu, C. H., Schneider, S. H., Fay, J. P., & Loarie, S. R. (2008). Cli-
mate change, elevational range shifts and bird extinctions. Conserva-
tion Biology, 22, 140–150. https://doi.org/10.1111/j.1523-1739.
2007.00852.x
Sirami, C., Caplat, P., Popy, S., Clamens, A., Arlettaz, R., Jiguet, F., … Mar-
tin, J.‐L. (2016). Impacts of global change on species distributions:
Obstacles and solutions to integrate climate and land use. Global Ecol-
ogy and Biogeography, 26, 385–394. https://doi.org/10.1111/geb.
12555
Smout, T. C. (2005). Nature contested. environmental history in Scotland
and Northern England since 1600. Edinburgh, Scotland: Edinburgh
University Press.
Soldaat, L. L., Pannekoek, J., Verweij, R. J. T., van Turnhout, C. A. M., &
van Strien, A. J. (2017). A Monte Carlo method to account for sam-
pling error in multi‐species indicators. Ecological Indicators, 81, 340–347. https://doi.org/10.1016/j.ecolind.2017.05.033
Stephens, P. A., Mason, L. R., Green, R. E., Gregory, R. D., Sauer, J. R.,
Alison, J., … Willis, S. G. (2016). Consistent response of bird popula-
tions to climate change on two continents. Science, 352(6281), 84–87. https://doi.org/10.1126/science.aac4858
Thompson, D. B. A., Kålås, J. A. K., & Byrkjedal, I. (2012). Arctic‐alpinemountain birds in northern Europe: Contrasts between specialists
and generalists. In R. J. Fuller (Ed.), Birds and habitats: Relationships in
changing landscapes (pp. 237–252). Cambridge, UK: Cambridge
University Press.
Thompson, D. B. A., MacDonald, A. J., Marsden, J. H., & Galbraith, C. A.
(1995). Upland heather moorland in Great Britain: A review of inter-
national importance, vegetation change and some objectives for nat-
ure conservation. Biological Conservation, 71, 163–178. https://doi.
org/10.1016/0006-3207(94)00043-P
Travis, J. M. J. (2003). Climate change and habitat destruction: a deadly
anthropogenic cocktail. Proceedings of the Royal Society B: Biological
Sciences, 270, 467–473.Turchin, P., Oksanen, L., Ekerholm, P., Oksanen, T., & Henttonen, H.
(2000). Are lemmings prey or predators? Nature, 405, 562–565.van der Wal, R., Pearce, I., Brooker, R., Scott, D., Welch, D., & Woodin,
S. (2003). Interplay between nitrogen deposition and grazing causes
habitat degradation. Ecology Letters, 6, 141–146. https://doi.org/10.1046/j.1461-0248.2003.00407.x
Vickery, J. A., Ewing, S. R., Smith, K. W., Pain, D. J., Bairlein, F., Škor-pilová, J., & Gregory, R. D. (2014). The decline of Afro‐Palaearcticmigrants and an assessment of potential causes. Ibis, 156, 1–22.https://doi.org/10.1111/ibi.12118
Vuorinen, K. E. M., Oksanen, L., Oksanen, T., Pyykönen, A., Olofsson, J.,
& Virtanen, R. (2017). Open tundra persists, but arctic features
decline – Vegetation changes in the warming Fennoscandian tundra.
Global Change Biology, 23, 3794–3807. https://doi.org/10.1111/gcb.13710
Zbinden, N., Kery, M., Keller, V., Brotons, L., Herrando, S., & Schmid, H.
(2010). Species richness of breeding birds along the altitudinal gradi-
ent: An analysis of atlas databases from Switzerland and Catalonia
(NE Spain). In E. M. Spehn, & C. Körner (Eds.), Data mining for global
trends in mountain biodiversity (pp. 65–73). Boca Raton, FL: Taylor &
Francis.
SUPPORTING INFORMATION
Additional supporting information may be found online in the
Supporting Information section at the end of the article.
How to cite this article: Lehikoinen A, Brotons L, Calladine J,
et al. Declining population trends of European mountain
birds. Glob Change Biol. 2018;00:1–12. https://doi.org/10.1111/gcb.14522
12 | LEHIKOINEN ET AL.